't think newt gingrich expected to be speaker in 94. neil: no, there is historic precedent to get it wrong, i don't know about you i will say, that in some states, and still early, as you remind me, it is beyond the margin for sampling error, i mean, if barack obama is up 8 to 10 points, even allowing for whatever statistical bias you see, mitt romney should be worried? >> he should be, but it does not make sense, nationally the race is statistically tied. you see the battle ground states we hear you know obama is so far ahead, that does not make sense, they are probably go 25 or 20% of the vote, in is something off right there i think that have you look at maybe 2010 models again, it was unexpected. neil: if did you, that by the way, the counter to that argument would be in corestates that are say republican states texas comes to mind, southern states, mitt romney enjoys significant leads, s is typically the case. corestates to each candidate go their way, ago -- having said that do you think that there is room for a wide swing within the swing states? if so, how significant would it be?